Method and device for determining effective computing power of server, equipment, medium and product
By obtaining the actual operating parameters of the server's core computing power components, its effective computing power is determined and standardized, solving the problem of inaccurate determination of server effective computing power in existing technologies, and realizing multi-dimensional accurate assessment and resource saving.
Patent Information
- Application Number
- CN202411072006.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-22
AI Technical Summary
In existing technologies, the methods for determining the effective computing power of servers are limited by the scientific nature and diversity of test sets, business scenarios, and software loads, and cannot fully represent the essential situation of the effective computing power of nodes, resulting in inaccurate determination results.
By obtaining the actual operating parameters of each core computing power component of the server, including dynamic and static operating parameters, the effective computing power of each core computing power component is determined, and based on the standardized effective computing power unified dimension, the actual effective computing power of the server is determined through fusion processing.
It enables accurate evaluation of server computing power performance from multiple dimensions, solves the problem of inaccurate determination caused by a single computing power component, and avoids the occupation of server resources during the testing process.
Smart Images

Figure CN119003303A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, device, equipment, medium and product for determining the effective computing power of a server. Background Art
[0002] At present, computing power is a new type of productivity in the digital economy era. Computing power is mainly provided by servers. The effective computing power of a server refers to the real computing power generated by the server during operation, that is, the actual computing performance. In evaluating the energy efficiency of the server (such as the carbon efficiency of computing power), the perception and scheduling of computing power resources, etc., it is necessary to measure the effective computing power of the server.
[0003] In the related art, the effective computing power of the server is usually determined by testing the server. This test method uses the actual throughput and performance of the test node using the test set, business scenario, and software load test, and normalizes it with the throughput benchmark value and the performance standard value, and uses the normalized indicator value to represent the effective computing power of the node. However, this test method is limited by the scientificity and diversity of the test set, business scenario, and software load. Therefore, this method cannot fully represent the essential situation of the effective computing power of the node. Therefore, the related art has great limitations in actual use. Summary of the invention
[0004] In view of the above problems, the present disclosure is proposed. The present disclosure provides a method, device, equipment, medium and product for determining effective computing power of a server.
[0005] According to one aspect of the present disclosure, a method for determining effective computing power of a server is provided, comprising:
[0006] Acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters;
[0007] Determine the effective computing power of each of the core computing power components based on the actual operating parameters;
[0008] Based on the effective computing power of each of the core computing power components, the normalized effective computing power of the core computing power component is determined, and based on the normalized effective computing power of all the core computing power components, the actual effective computing power of the server is determined.
[0009] In addition, according to another embodiment of one aspect of the present disclosure, the determining the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components includes:
[0010] Determining operating parameters of each benchmark computing component of the same type as the core computing component in the benchmark server;
[0011] Determine the nominal computing power of each of the benchmark computing power components based on the operating parameters;
[0012] The normalized effective computing power of the core computing power component is determined according to the nominal computing power and the effective computing power of the core computing power component.
[0013] In addition, according to another embodiment of one aspect of the present disclosure, determining the actual effective computing power of the server based on the normalized effective computing power of all the core computing power components includes:
[0014] Based on the normalized effective computing power of each of the core computing power components, determine the normalized effective computing power of the server; wherein the normalized effective computing power of the server is used to indicate the multiple relationship between the effective computing power of the server and the nominal computing power of the benchmark server;
[0015] Based on the normalized effective computing power of the server, the actual effective computing power of the server is determined.
[0016] In addition, according to another embodiment of one aspect of the present disclosure, the core computing power component includes at least one of the following: a central processing unit, a graphics processing unit, a memory, and a hard disk.
[0017] In addition, according to another embodiment of one aspect of the present disclosure, the determining the normalized effective computing power of the server based on the normalized effective computing power of each of the core computing power components includes:
[0018] Performing weighted summation on the normalized effective computing power of the central processing unit and the normalized effective computing power of the graphics processing unit according to the multiplication factor to obtain a weighted summation result;
[0019] Determine the weight coefficient of the weighted summation result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk; wherein the weight coefficient is used to indicate the importance of the corresponding core computing power component to the server;
[0020] The weighted summation result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk are subjected to exponential multiplication according to the weight coefficient to obtain a calculation result, and the normalized effective computing power of the server is determined according to the calculation result.
[0021] In addition, according to another embodiment of one aspect of the present disclosure, the core computing power component is a central processing unit; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes:
[0022] Acquire a first dynamic operating parameter and a first static operating parameter of the central processor; wherein the first static operating parameter includes the total number of cores of the central processor of the server and the single-cycle floating-point computing capability of each core of the central processor of the server, and the first dynamic operating parameter includes the single-core main frequency of each core of the central processor of the server and the utilization rate of each core of the central processor of the server;
[0023] The effective computing power of the central processing unit is calculated according to the first dynamic operating parameter and the first static operating parameter.
[0024] In addition, according to another embodiment of one aspect of the present disclosure, the core computing power component is a graphics processor; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes:
[0025] Acquire a second dynamic operating parameter and a second static operating parameter of the graphics processor; wherein the second static operating parameter includes the total number of cores of the graphics processor of the server and the single-cycle floating-point computing capability of each core of the graphics processor of the server, and the second dynamic operating parameter includes the single-core main frequency of each core of the graphics processor of the server and the utilization rate of each core of the graphics processor of the server;
[0026] The effective computing power of the graphics processor is calculated according to the second dynamic operating parameter and the second static operating parameter.
[0027] In addition, according to another embodiment of one aspect of the present disclosure, the core computing power component is a memory; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes:
[0028] Obtaining a third dynamic operating parameter and a third static operating parameter of the memory; wherein the third static operating parameter includes the number of memory bars of the server, the memory capacity of each memory bar of the server, the memory operating frequency of the server, the number of memory channels of the server, the bit width of the memory channel of the server, and the conversion coefficient of the memory configuration of each memory channel of the server to the memory computing power; and the third dynamic operating parameter includes the memory usage rate of each memory bar of the server;
[0029] The effective computing power of the memory is calculated according to the third dynamic operating parameter and the third static operating parameter.
[0030] In addition, according to another embodiment of one aspect of the present disclosure, the core computing power component is a hard disk; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes:
[0031] Acquire a fourth dynamic operating parameter and a fourth static operating parameter of the hard disk; wherein the fourth static operating parameter includes the number of hard disks of the server and the hard disk capacity of each hard disk of the server, and the fourth dynamic operating parameter includes the read and write speed of each hard disk of the server and the utilization rate of each hard disk of the server;
[0032] The effective computing power of the hard disk is calculated according to the fourth dynamic operating parameter and the fourth static operating parameter.
[0033] According to another aspect of the present disclosure, a device for determining effective computing power of a server is provided, comprising:
[0034] An acquisition module, used to acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters;
[0035] A first determination module, configured to determine the effective computing power of each of the core computing power components based on the actual operating parameters;
[0036] The second determination module is used to determine the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components, and to determine the actual effective computing power of the server based on the normalized effective computing power of all the core computing power components.
[0037] According to another aspect of the present disclosure, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement steps of a method for determining a log template.
[0038] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the steps of a method for determining a log template are implemented.
[0039] According to another aspect of the present disclosure, a computer program product is provided, including a computer program / instruction, which implements the steps of a method for determining a log template when the computer program / instruction is executed by a processor.
[0040] As will be described in detail below, according to the method, device, equipment, medium and product for determining the effective computing power of the server in the embodiments of the present disclosure. The technical solution disclosed in the present disclosure, by determining the effective computing power of each core component of the server, and then determining the normalized effective computing power of each core component based on the effective computing power, can unify the dimensions of the effective computing power of each core computing power component, and then can perform fusion processing based on the effective computing power of the unified dimension, thereby obtaining the actual effective computing power that reflects the computing power performance of the server from multiple dimensions. This processing method can solve the problem of inaccurate determination of the actual effective computing power of the server due to the single selected computing power component in the prior art.
[0041] At the same time, the technical solution disclosed in the present invention can solve the problem of inaccurate determination of the actual effective computing power of the server due to the difference between the test set and the actual tasks performed by the server by collecting the actual operating parameters generated during the actual operation of the server to determine the actual effective computing power of the server. At the same time, it can also avoid the occupation of server resources during the test process.
[0042] It is to be understood that both the foregoing general description and the following detailed description are exemplary, and are intended to provide further explanation of the technology as claimed. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the accompanying drawings, the same reference numerals generally represent the same components or steps.
[0044] Figure 1 A flowchart of a method for determining the effective computing power of a server provided in an embodiment of the present disclosure.
[0045] Figure 2 A specific flow chart of the method for determining the effective computing power of a server provided in an embodiment of the present disclosure.
[0046] Figure 3 A schematic diagram of a device for determining the effective computing power of a server provided in an embodiment of the present disclosure.
[0047] Figure 4 A schematic diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION
[0048] In order to make the purpose, technical solution and advantages of the present disclosure more obvious, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.
[0049] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0050] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.
[0051] Research has found that computing power is a new type of productivity in the digital economy era. Computing power is mainly provided by servers. The effective computing power of a server refers to the real computing power generated by the server during operation, that is, the actual computing performance. In evaluating the energy efficiency of the server (such as the carbon efficiency of computing power), the perception and scheduling of computing power resources, etc., it is necessary to measure the effective computing power of the server.
[0052] In the related art, the effective computing power of the server is usually determined by testing the server. This test method uses the actual throughput and performance of the test node using the test set, business scenario, and software load test, and normalizes it with the throughput benchmark value and the performance standard value, and uses the normalized indicator value to represent the effective computing power of the node. However, this test method is limited by the scientificity and diversity of the test set, business scenario, and software load. Therefore, this method cannot fully represent the essential situation of the effective computing power of the node. Therefore, the related art has great limitations in actual use.
[0053] Based on the above research, the present disclosure provides a method for determining the effective computing power of a server.
[0054] By determining the effective computing power of each core component of the server, and then determining the normalized effective computing power of each core component based on the effective computing power, the dimensions of the effective computing power of each core computing power component can be unified, and then the effective computing power of the unified dimension can be integrated and processed to obtain the actual effective computing power that reflects the computing power performance of the server from multiple dimensions. This processing method can solve the problem of inaccurate determination of the actual effective computing power of the server due to the single selected computing power component in the prior art.
[0055] At the same time, the technical solution disclosed in the present invention can solve the problem of inaccurate determination of the actual effective computing power of the server due to the difference between the test set and the actual tasks performed by the server by collecting the actual operating parameters generated during the actual operation of the server to determine the actual effective computing power of the server. At the same time, it can also avoid the occupation of server resources during the test process.
[0056] To facilitate understanding of this embodiment, a method for determining the effective computing power of a server disclosed in an embodiment of the present disclosure is first introduced in detail. The execution subject of the method for determining the effective computing power of a server provided in an embodiment of the present disclosure is generally an electronic device with certain computing capabilities. In some possible implementations, the method for determining the effective computing power of a server can be implemented by a processor calling computer-readable instructions stored in a memory.
[0057] Figure 1 FIG. 1 is a flowchart of a method for determining effective computing power of a server provided in an embodiment of the present disclosure, wherein the method includes steps S101 to S103, wherein:
[0058] S101. Obtain actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters.
[0059] In the embodiment of the present disclosure, the actual operation parameters are used to indicate the parameters generated during the actual operation of the server when executing various computing tasks. Here, the actual operation parameters of the server can be collected according to a preset time interval.
[0060] Here, the dynamic working parameters are used to indicate the parameters that are dynamically changing during the working process of the server, and the static working parameters are used to indicate the parameters that are fixed and unchanged during the working process of the server.
[0061] Here, the core computing power components may include at least one of the following: a central processing unit, a graphics processing unit, a memory, and a hard disk.
[0062] S102. Determine the effective computing power of each core computing power component based on actual operating parameters.
[0063] In the embodiments of the present disclosure, the operating parameters corresponding to each core computing power component can be determined in the actual operating parameters, and the operating parameters can be calculated to obtain the effective computing power (Effective Computing Power, ECP for short) of the core computing power component.
[0064] Here, the dimensions corresponding to the effective computing power of each core computing power component may be different. For example, when the core computing power component is a central processing unit (CPU) or a graphics processing unit (GPU), the dimension of the effective computing power of the core computing power component may be floating-point operations per second (flops); when the core computing power component is a memory, the dimension of the effective computing power of the core computing power component may be bytes·bits per second (i.e., B·bps, capacity and data transmission bandwidth); when the core computing power component is a hard disk, the dimension of the effective computing power of the core computing power component may be bytes·bits per second (i.e., B·bps, capacity and data transmission bandwidth).
[0065] S103. Based on the effective computing power of each core computing power component, determine the normalized effective computing power of the core computing power component, and based on the normalized effective computing power of all core computing power components, determine the actual effective computing power of the server.
[0066] In an embodiment of the present disclosure, a reference server corresponding to a server and ideal operating parameters of each reference computing component of the reference server can be determined, wherein the ideal operating parameters are operating parameters of each reference computing component of the reference server under an ideal operating state.
[0067] For example, it can be determined when each benchmark computing power component is in an ideal operating state, or it can be determined by referring to the component description of each benchmark computing power component.
[0068] Here, the Normalized Effective Computing Power (NECP) of each core computing component can be determined based on the actual operating parameters and the ideal operating parameters.
[0069] Here, for each core computing power component, the normalized effective computing power of the core computing power component can be determined based on the actual operating parameters corresponding to the core computing power component and the ideal operating parameters corresponding to the benchmark computing power component that matches the core computing power component.
[0070] Here, the normalized effective computing power of each core computing power component can be integrated and calculated to obtain the normalized effective computing power of the server.
[0071] In the embodiments of the present disclosure, first, the actual operating parameters of each core computing power component of the server during the actual operation are obtained; wherein the actual operating parameters include dynamic operating parameters and static operating parameters; secondly, the effective computing power of each core computing power component is determined based on the actual operating parameters; finally, based on the effective computing power of each core computing power component, the normalized effective computing power of the core computing power component is determined, and based on the normalized effective computing power of all core computing power components, the actual effective computing power of the server is determined.
[0072] The above implementation method, by determining the effective computing power of each core component of the server, and then determining the normalized effective computing power of each core component based on the effective computing power, can unify the dimensions of the effective computing power of each core computing power component, and then perform fusion processing based on the effective computing power of the unified dimension, thereby obtaining the actual effective computing power that reflects the computing power performance of the server from multiple dimensions. This processing method can solve the problem of inaccurate determination of the actual effective computing power of the server due to the single selected computing power component in the prior art.
[0073] At the same time, the technical solution disclosed in the present invention can solve the problem of inaccurate determination of the actual effective computing power of the server due to the difference between the test set and the actual tasks performed by the server by collecting the actual operating parameters generated during the actual operation of the server to determine the actual effective computing power of the server. At the same time, it can also avoid the occupation of server resources during the test process.
[0074] In an optional implementation, the above steps determine the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components, specifically including the following steps:
[0075] First, determining the operating parameters of each benchmark computing component of the same type as the core computing component in the benchmark server;
[0076] Secondly, determining the nominal computing power of each of the benchmark computing power components based on the operating parameters;
[0077] Finally, the normalized effective computing power of the core computing power component is determined based on the nominal computing power and the effective computing power of the core computing power component.
[0078] In the embodiments of the present disclosure, the component types of the core computing power components include, but are not limited to: processor, memory, hard disk. Among them, the processor includes: central processing unit and image processing unit.
[0079] Here, first, the component type of the core computing power component can be determined. Secondly, the computing power component of the same type can be determined as the benchmark computing power component in the benchmark server according to the component type of the core computing power component. Finally, the operating parameters of each benchmark computing power component can be determined. At this time, the operating parameters are the ideal operating parameters described in the above content.
[0080] Here, the operating parameters of each benchmark computing power component include dynamic operating parameters and static operating parameters.
[0081] Here, the operating parameters of each benchmark computing power component can be calculated to obtain the nominal computing power (Computing Power, CP for short) of the benchmark computing power component.
[0082] Here, the effective computing power of each type of core computing power component and the nominal computing power of the benchmark computing power component that matches the core computing power component can be calculated separately to obtain the normalized effective computing power of the core computing power component.
[0083] In an optional implementation, the above steps determine the actual effective computing power of the server based on the normalized effective computing power of all core computing power components, and specifically include the following steps:
[0084] First, based on the normalized effective computing power of each core computing power component, the normalized effective computing power of the server is determined; wherein the normalized effective computing power of the server is used to indicate the multiple relationship between the effective computing power of the server and the nominal computing power of the benchmark server;
[0085] Then, based on the normalized effective computing power of the server, the actual effective computing power of the server is determined.
[0086] In an embodiment of the present disclosure, the core computing power component includes at least one of the following: a central processing unit, a graphics processing unit, a memory, and a hard disk.
[0087] Specifically, based on the normalized effective computing power of each core computing power component, the normalized effective computing power of the server is determined, which specifically includes the following steps:
[0088] First, the normalized effective computing power of the central processing unit and the normalized effective computing power of the graphics processing unit are weighted and summed according to the multiplication coefficient to obtain a weighted sum result;
[0089] Secondly, determine the weight coefficients of the weighted summation result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk; wherein the weight coefficient is used to indicate the importance of the corresponding core computing power component to the server;
[0090] Finally, the weighted summation result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk are exponentially multiplied according to the weight coefficient to obtain the calculation result, and the normalized effective computing power of the server is determined based on the calculation result.
[0091] Here, each core computing component can be distinguished by type. Among them, the central processing unit and the graphics processing unit belong to the same type and use the same dimension.
[0092] Here, the multiplication factor can be determined based on the nominal computing power of the CPU and the nominal computing power of the GPU of the reference server. After the multiplication factor is determined, the normalized effective computing power of the CPU and the normalized effective computing power of the GPU can be weighted and summed according to the multiplication factor so that the effective computing power of the CPU and the effective computing power of the GPU can be equivalently converted.
[0093] Here, the multiplication factor α meets the following conditions:
[0094] α=CP 基准GPU / CP 基准CPU ;
[0095] Among them, CP 基准GPU is the nominal computing power of the GPU of the benchmark server, CP 基准CPU The nominal computing power of the CPU of the benchmark server.
[0096] Here, the normalized effective computing power of the processor can be determined by the following steps. First, the effective computing power of the central processing unit and the graphics processing unit of the server can be determined. Secondly, the nominal computing power of the central processing unit and the graphics processing unit of the benchmark server can be determined. Secondly, the effective computing power of the central processing unit and the nominal computing power can be divided to obtain a first division result. Secondly, the effective computing power of the graphics processor and the nominal computing power can be divided to obtain a second division result. Finally, the first division result and the second division result can be weighted and summed based on the multiplication factor to obtain the normalized effective computing power of the processor.
[0097] Here, the adjustment item of the processor may be determined based on the multiplication factor, wherein the adjustment item is used to adjust the normalized effective computing power value of the processor to 1, so that the normalized effective computing power value of the processor is more consistent with the objective physical meaning and law.
[0098] Here, the sum of the weighted summation result, the normalized effective computing power of the memory, and the weight coefficient of the normalized effective computing power of the hard disk is 1. The weight coefficient of the weighted summation result is the weight coefficient of the processor in the server; the weight coefficient of the normalized effective computing power of the memory is the weight coefficient of the memory; the weight coefficient of the normalized effective computing power of the hard disk is the weight coefficient of the hard disk.
[0099] Here, the weight coefficients can be set empirically. For example, the weight coefficient e1 of the weighted summation result, the weight coefficient e2 of the normalized effective computing power of the memory, and the weight coefficient e3 of the normalized effective computing power of the hard disk can be set to 0.65, 0.3, and 0.05, respectively.
[0100] Here, the weight coefficient e1 of the weighted summation result can be subjected to exponential operation with the weighted summation result to obtain a first exponential result. The weight coefficient e2 of the normalized effective computing power of the memory can be subjected to exponential operation with the normalized effective computing power of the memory to obtain a second exponential result. The weight coefficient e3 of the normalized effective computing power of the hard disk can be subjected to exponential operation with the normalized effective computing power of the hard disk to obtain a third exponential result.
[0101] Here, the first index result, the second index result and the third index result may be multiplied to obtain a calculation result.
[0102] Here, after obtaining the calculation result, the division between the calculation result and the adjustment item can be calculated to obtain the normalized effective computing power of the server. Here, the normalized effective computing power NECP of the server meets the following conditions:
[0103] NECP=(ECP CPU / CP 基准CPU +α*ECP GPU / CP 基准GPU ) e1 *(ECP 内存 / CP 基准内
[0104] 存 ) e2 *(ECP 硬盘 / CP 基准硬盘 ) e3 / (1+α) e1 ;
[0105] Among them, ECP CPU The effective computing power of the server’s central processing unit, CP 基准CPU is the nominal computing power of the CPU of the benchmark server, α is the multiplication factor, and ECP GPU is the effective computing power of the server’s graphics processor, CP 基准GPU is the nominal computing power of the GPU of the benchmark server, ECP 内存 is the effective computing power of the server's memory, CP 基准内存 is the nominal computing power of the benchmark server’s memory, ECP 硬盘 is the effective computing power of the server's hard disk, CP 基准硬盘 is the nominal computing power of the hard disk of the benchmark server, (1+α) e1 For adjustment items.
[0106] Here, after determining the normalized effective computing power of the server, the ideal computing power of the reference server can be determined, and the actual effective computing power of the server can be determined based on the ideal computing power of the reference server and the normalized effective computing power of the server. For example, the ideal computing power of the reference server and the normalized effective computing power of the server are multiplied to obtain the actual effective computing power of the server.
[0107] In an optional implementation, when the core computing power component is a central processing unit, the above steps determine the effective computing power of each core computing power component based on the actual operating parameters, specifically including the following steps:
[0108] First, a first dynamic operating parameter and a first static operating parameter of the central processing unit are obtained; wherein the first static operating parameter includes the total number of cores of the central processing unit of the server and the single-cycle floating-point computing capability of each core of the central processing unit of the server, and the first dynamic operating parameter includes the single-core main frequency (i.e., average operating frequency) of each core of the central processing unit of the server and the utilization rate of each core of the central processing unit of the server;
[0109] Then, the effective computing power of the central processing unit is calculated according to the first dynamic operating parameter and the first static operating parameter.
[0110] In an embodiment of the present disclosure, the core frequency of each core of the central processing unit, the single-cycle floating-point computing capability corresponding to the core, and the utilization rate of the core may be multiplied to obtain the effective computing power of the core.
[0111] Here, after determining the effective computing power of each core of the central processing unit of the server, the effective computing power of each core may be summed to obtain the effective computing power of the central processing unit.
[0112] Among them, the effective computing power of the central processing unit ECP CPU Meet the following conditions:
[0113] ECP CPU =∑ N(CPU核) i=1 (F CPU(i) *FP CPU(i) *UR CPU(i) );
[0114] Among them, F CPU(i) is the single core frequency of the i-th core of the server's CPU, i ranges from 1 to N (CPU cores), N (CPU cores) is the total number of cores of the server's CPU, FP CPU(i) is the single-cycle floating-point computing capability of the i-th core of the server’s CPU, UR CPU(i) is the usage rate of the i-th core of the server's CPU.
[0115] For example, the configuration of the central processing unit of the server is 2×Intel(R)Xeon(R)Gold 6238CPU@2.10GHz. It can be seen that the total number of cores of the central processing unit of the server in the first static working parameter is 2(CPU number)×22(single CPU core number)=44.
[0116] Here, the single-cycle floating-point computing capability of each core of the server's central processing unit supports up to the AVX-512 instruction set, with an instruction length of 512 bits. Each CPU core has two FMA (multi-mode floating-point fused multiply-add) units. Each FMA unit performs one floating-point addition and one floating-point multiplication operation in one clock cycle. Calculated according to single-precision floating-point numbers (FP32), the single-cycle floating-point computing capability of each core of the central processing unit is 512 / 32×2×2=64 times.
[0117] Here, the single core main frequency of each core of the central processor of the server in the first dynamic working parameter is 2.1 GHz; the utilization rate of each core of the central processor of the server can be collected through the virtualization platform, for example, 15%.
[0118] Here, the nominal computing power of the central processing unit of the benchmark server can be calculated in the above manner.
[0119] Among them, the nominal computing power CP of the central processing unit 基准CPU Meet the following conditions:
[0120] CP 基准CPU =∑ N’(CPU核) j=1 (F' CPU(j) *FP' CPU(j) );
[0121] Among them, F' CPU(j) is the single-core nominal main frequency of the jth core of the CPU of the benchmark server, and the value of j ranges from 1 to N' (CPU core), where N' (CPU core) is the total number of cores of the CPU of the benchmark server, and FP' CPU(j) It is the single-cycle floating-point computing capability of the i-th core of the CPU of the benchmark server.
[0122] For example, the CPU configuration of the benchmark server is 1×Intel Celeron CPU G1101@2.26GHz. It can be seen that the total number of cores of the benchmark server's CPU is 2; the single-core nominal main frequency of each core of the benchmark server's CPU is 2.26GHz; and the single-cycle floating-point computing capability of each core of the benchmark server's CPU is 2 times.
[0123] Here, we determine the effective computing power (ECP) of the server's CPU. CPU and the nominal computing power CP of the CPU of the benchmark server 基准CPU After that, the effective computing power of the server’s CPU ECP can be CPU Divide by the nominal CPU power of the benchmark server 基准CPU , get the normalized effective computing power of the server’s CPU.
[0124] In an optional implementation, when the core computing power component is a graphics processor, determining the effective computing power of each core computing power component based on actual operating parameters specifically includes the following steps:
[0125] First, a second dynamic operating parameter and a second static operating parameter of the graphics processor are obtained; wherein the second static operating parameter includes the total number of cores of the graphics processor of the server and the single-cycle floating-point computing capability of each core of the graphics processor of the server, and the second dynamic operating parameter includes the single-core main frequency (i.e., the average working main frequency) of each core of the graphics processor of the server and the utilization rate of each core of the graphics processor of the server;
[0126] Secondly, the effective computing power of the graphics processor is calculated according to the second dynamic operating parameter and the second static operating parameter.
[0127] In an embodiment of the present disclosure, the core frequency of each core of the graphics processor, the single-cycle floating-point computing capability corresponding to the core, and the usage rate of the core may be multiplied to obtain the effective computing power of the core.
[0128] Here, after determining the effective computing power of each core of the graphics processor of the server, the effective computing power of each core may be summed to obtain the effective computing power of the graphics processor.
[0129] Among them, the effective computing power of the graphics processor ECP GPU Meet the following conditions:
[0130] ECP GPU =∑ N(GPU核) d=1 (F GPU(d) *FP GPU(d) *UR GPU(d) );
[0131] Among them, F GPU(d) is the single core frequency of the dth core of the server's CPU, d ranges from 1 to N (GPU core), N (GPU core) is the total number of cores of the server's GPU, FP GPU(d) is the single-cycle floating-point computing capability of the dth core of the server’s graphics processor, UR GPU(d)is the usage rate of the dth core of the server's graphics processor.
[0132] For example, the configuration of the graphics processor of the server is 2×NVIDIA RTX A6000. It can be seen that the total number of CUDA (Compute Unified Device Architecture) cores of the graphics processor in the second static working parameter is 2 (i.e., the number of GPUs) × 10752 (i.e., the number of CUDA cores of a single GPU) = 21504, and the total number of Tensor cores is 2 (i.e., the number of GPUs) × 336 (i.e., the number of Tensor cores of a single GPU) = 672; the single-cycle floating-point computing capability of each CUDA core of the graphics processor of the server is 2 times, and the single-cycle floating-point computing capability of each Tensor core is 256 times.
[0133] Here, the single-core main frequency of the graphics processor of the server in the second dynamic working parameter is 1.455 GHz; the utilization rate of each core of the graphics processor of the server can be obtained by querying the CUDA and Tensor core utilization rate using the nvidia-smi command, for example, 30%.
[0134] Here, the nominal computing power of the graphics processor of the benchmark server may be calculated in the above manner.
[0135] Among them, the nominal computing power CP of the graphics processor 基准GPU Meet the following conditions:
[0136] CP 基准GPU =∑ N’(GPU核) f=1 (F' GPU(f) *FP' GPU(f) );
[0137] Among them, F' GPU(f) is the single core frequency of the fth core of the GPU of the benchmark server, where f ranges from 1 to N'(GPU core), where N'(GPU core) is the total number of cores of the GPU of the benchmark server, and FP' GPU(f) The single-cycle floating-point computing capability of the f-th core of a benchmark server's graphics processor.
[0138] For example, the configuration of the graphics processor of the benchmark server is 1×NVIDIA Tesla V100@1.38GHz PCIeGEN3. It can be seen that the total number of CUDA cores of the graphics processor of the benchmark server is 1 (number of GPUs)×5120 (number of CUDA cores of a single GPU)=5120, and the total number of Tensor cores is 640; the single-core nominal main frequency of each core of the graphics processor of the benchmark server is 1.38GHz; the single-cycle floating-point computing capability of each CUDA core of the graphics processor of the benchmark server is 2 times, and the single-cycle floating-point computing capability of each Tensor core is 64 times.
[0139] Here, we determine the effective computing power (ECP) of the server's graphics processor. GPU The nominal computing power CP of the GPU of the benchmark server 基准GPU After that, the effective computing power ECP of the server's graphics processor can be GPU Divide by the nominal computing power CP of the benchmark server's graphics processor 基准GPU , get the normalized effective computing power of the server’s graphics processor.
[0140] In an optional implementation, when the core computing power component is a memory, determining the effective computing power of each core computing power component based on actual operating parameters specifically includes the following steps:
[0141] First, obtain a third dynamic working parameter and a third static working parameter of the memory; wherein the third static working parameter includes the number of memory bars of the server, the memory capacity of each memory bar of the server, the memory working frequency of the server, the number of memory channels of the server, the bit width of the memory channel of the server, and the conversion coefficient of the memory configuration of each memory channel of the server to the memory computing power; and the third dynamic working parameter includes the memory usage rate of each memory bar of the server;
[0142] Then, the effective computing power of the memory is calculated according to the third dynamic operating parameter and the third static operating parameter.
[0143] In an embodiment of the present disclosure, the memory capacity of each memory bar of the memory and the usage rate of the memory bar may be multiplied to obtain the used memory capacity of the memory bar.
[0144] Here, after determining the used memory capacity of each memory bar of the server's memory, the used memory capacity of each memory bar may be summed to obtain the overall used memory capacity of the memory.
[0145] Here, after obtaining the overall used memory capacity of the memory, the overall used memory capacity of the memory, the memory operating frequency of the server, the number of memory channels of the server, the memory channel width of the server, and the memory configuration of each memory channel of the server (DIMM Per Channel, DPC for short) can be multiplied by the conversion coefficient of the memory computing power to obtain the effective computing power of the server's memory.
[0146] Among them, the effective computing power of memory ECP 内存 Meet the following conditions:
[0147] ECP 内存 =(∑ N(内存) s=1 (C 内存(s) *UR 内存(s) ))*F 内存 *CH 内存 *W 内存CH *DPC
[0148] 内存 *2;
[0149] Among them, C 内存(s) is the memory capacity of the sth memory bank of the server, s ranges from 1 to N (memory), N (memory) is the number of memory banks of the server, UR 内存(s) is the memory usage of the sth memory bank of the server, F 内存 is the memory operating frequency of the server, CH 内存 is the number of memory channels of the server, W 内存CH is the memory channel width of the server, DPC 内存 This is the conversion factor of the memory configuration of each memory channel of the server to the memory computing power.
[0150] For example, the memory configuration of the server is 12×Micron 32GB DDR4-2933 ECC RDIMM. It can be seen that the number of memory bars of the server in the third static working parameter is 12, the memory capacity of each memory bar of the server is 32GB, the memory working frequency of the server is 2.933GHz, the number of memory channels of the server is 6, the bit width of the memory channel of the server is 64it, and the conversion coefficient of the memory configuration of each memory channel of the server to the memory computing power is 100%.
[0151] Here, the third dynamic working parameter (ie, the memory usage rate of each memory bank of the server) can be collected and obtained through the virtualization platform, for example, 65%.
[0152] Here, the nominal computing power of the memory of the benchmark server can be calculated in the above manner.基准内存 Meet the following conditions:
[0153] CP 基准内存 =(∑ N’(内存) q=1 C' 内存(q) )*F' 内存 *CH' 内存 *W' 内存CH *DPC' 内存 *2;
[0154] Among them, C' 内存(q) is the memory capacity of the qth memory bank of the benchmark server, the value of q ranges from 1 to N'(memory), N'(memory) is the number of memory banks of the benchmark server, F' 内存 is the memory operating frequency of the benchmark server, CH' 内存 is the number of memory channels of the benchmark server, W' 内存CH is the memory channel width of the benchmark server, DPC' 内存 It is the conversion factor of memory computing power based on the memory configuration of each memory channel of the benchmark server.
[0155] For example, the memory configuration of the benchmark server is 4×2GB PC3L-10600E DIMMs, DDR3 1333MHz. It can be seen that the number of memory bars of the benchmark server is 4, the memory capacity of each memory bar of the benchmark server is 2GB, the memory operating frequency of the benchmark server is 1.066GHz (because the CPU of the benchmark server supports a maximum memory operating frequency of 1.066GHz, the memory is down-clocked to 1.066GHz), the number of memory channels of the benchmark server is 2, the memory channel width of the benchmark server is 64it, and the memory configuration of each memory channel of the benchmark server has a conversion coefficient of 95% for memory computing power.
[0156] Here, we determine the effective computing power ECP of the server's memory. 内存 The nominal computing power CP of the memory of the benchmark server 基准内存 After that, the effective computing power ECP of the server's memory can be 内存 Divide the nominal computing power CP by the memory of the benchmark server 基准内存 , get the normalized effective computing power of the server's memory.
[0157] In an optional implementation, the core computing power component is a hard disk; determining the effective computing power of each core computing power component based on the actual operating parameters specifically includes the following steps:
[0158] First, a fourth dynamic working parameter and a fourth static working parameter of the hard disk are obtained; wherein the fourth static working parameter includes the number of hard disks of the server and the hard disk capacity of each hard disk of the server, and the fourth dynamic working parameter includes the read and write speed of each hard disk of the server and the utilization rate of each hard disk of the server;
[0159] Then, the effective computing power of the hard disk is calculated according to the fourth dynamic operating parameter and the fourth static operating parameter.
[0160] In an embodiment of the present disclosure, the hard disk capacity of each hard disk, the read and write speed of the hard disk, and the utilization rate of the hard disk may be multiplied to obtain the effective computing power of the hard disk.
[0161] Here, after determining the effective computing power of each hard disk of the server, the effective computing power of each hard disk can be summed to obtain the effective computing power of the hard disk of the server. 硬盘 Meet the following conditions:
[0162] ECP 硬盘 =∑ N(硬盘) y=1 (C 硬盘(y) *S 硬盘(y) *UR 硬盘(y) );
[0163] Among them, C 硬盘(y) is the hard disk capacity of the yth hard disk in the server, y ranges from 1 to N (hard disk), N (hard disk) is the number of hard disks in the server, S 硬盘(y) is the read and write speed of the yth hard disk in the server, UR 硬盘(y) is the usage rate of the yth hard disk in the server.
[0164] For example, the configuration of the hard disk of the server is 3×SAMSUNG MZ7LH480 SSD 446.62GB SATA6.0Gbps. It can be seen that the number of hard disks of the server in the fourth static working parameter is 3; and the hard disk capacity of each hard disk of the server is 446.62GB.
[0165] Here, in the fourth dynamic working parameter, the read speed of each hard disk of the server is 712.86KB / s, the write speed is 76.66KB / s, and the read and write speeds are 712.86KB / s+76.66KB / s; the utilization rate of each hard disk of the server can be collected through the virtualization platform, for example, 45%.
[0166] Here, the above method can be used to calculate the nominal computing power of the hard disk of the reference server. 硬盘 Meet the following conditions:
[0167] CP 硬盘 =∑ N’(硬盘) h=1 (C' 硬盘(h) *S' 硬盘(h) );
[0168] Among them, C' 硬盘(h) is the hard disk capacity of the hth hard disk in the benchmark server, h ranges from 1 to N'(hard disk), N'(hard disk) is the number of hard disks in the benchmark server, S' 硬盘(h) is the nominal read and write speed of the hth hard disk in the benchmark server.
[0169] For example, the number of hard disks of the reference server is 1, the hard disk capacity of each hard disk of the reference server is 80 GB, and the nominal read / write speed of each hard disk of the reference server is 1.5 Gsbps.
[0170] Here, we determine the effective computing power ECP of the server's hard disk. 硬盘 And the nominal computing power CP of the hard disk of the benchmark server 基准硬盘 After that, the effective computing power ECP of the server's hard disk can be 硬盘 Divide by the nominal computing power CP of the hard disk of the benchmark server 基准硬盘 , get the normalized effective computing power of the server's hard disk.
[0171] See also Figure 2 As shown, it is a specific flow chart of the method for determining the effective computing power of a server provided in an embodiment of the present disclosure, wherein:
[0172] S10, obtaining actual operating parameters of the server's central processing unit, graphics processing unit, memory, and hard disk during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters.
[0173] S20, determining the effective computing power of the central processing unit, the graphics processing unit, the memory and the hard disk based on actual operating parameters of the central processing unit, the graphics processing unit, the memory and the hard disk of the server;
[0174] S30, determining static operating parameters of the central processing unit, graphics processing unit, memory and hard disk of the reference server.
[0175] S40: Determine the nominal computing power of the central processing unit, graphics processing unit, memory and hard disk of the benchmark server based on the static working parameters of the central processing unit, graphics processing unit, memory and hard disk of the benchmark server.
[0176] S50. Determine a multiplication factor and an adjustment item based on the nominal computing power of the central processing unit and the graphics processing unit of the benchmark server.
[0177] S60, determining a weighted sum result based on the effective computing power of the central processing unit and the graphics processing unit of the server, the nominal computing power of the central processing unit and the graphics processing unit of the benchmark server, the multiplication factor and the adjustment item.
[0178] S70. Determine the normalized effective computing power of the server's memory and hard disk based on the effective computing power of the server's memory, the effective computing power of the hard disk, and the nominal computing power of the memory and the nominal computing power of the hard disk of the benchmark server.
[0179] S80. Determine the normalized effective computing power of the server based on the weighted sum result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk.
[0180] S90. Determine the actual effective computing power of the server based on the normalized effective computing power of the server.
[0181] The above implementation method, by determining the effective computing power of each core component of the server, and then determining the normalized effective computing power of each core component based on the effective computing power, can unify the dimensions of the effective computing power of each core computing power component, and then perform fusion processing based on the effective computing power of the unified dimension, thereby obtaining the actual effective computing power that reflects the computing power performance of the server from multiple dimensions. This processing method can solve the problem of inaccurate determination of the actual effective computing power of the server due to the single selected computing power component in the prior art.
[0182] At the same time, the technical solution disclosed in the present invention can solve the problem of inaccurate determination of the actual effective computing power of the server due to the difference between the test set and the actual tasks performed by the server by collecting the actual operating parameters generated during the actual operation of the server to determine the actual effective computing power of the server. At the same time, it can also avoid the occupation of server resources during the test process.
[0183] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0184] Based on the same inventive concept, the embodiments of the present disclosure also provide a device for determining the effective computing power of a server corresponding to the method for determining the effective computing power of a server. Since the principle of solving the problem by the device in the embodiments of the present disclosure is similar to the method for determining the effective computing power of a server in the embodiments of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0185] See also Figure 3 FIG. 1 is a schematic diagram of a device for determining effective computing power of a server provided by an embodiment of the present disclosure, wherein the device comprises: an acquisition module 11, a first determination module 12, and a second determination module 13; wherein:
[0186] An acquisition module, used to acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters;
[0187] A first determination module, configured to determine the effective computing power of each of the core computing power components based on the actual operating parameters;
[0188] The second determination module is used to determine the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components, and to determine the actual effective computing power of the server based on the normalized effective computing power of all the core computing power components.
[0189] The disclosed embodiment determines the effective computing power of each core component of the server, and then determines the normalized effective computing power of each core component based on the effective computing power, so as to unify the dimensions of the effective computing power of each core computing power component, and then perform fusion processing based on the effective computing power of the unified dimension, thereby obtaining the actual effective computing power that reflects the computing power performance of the server from multiple dimensions. This processing method can solve the problem of inaccurate determination of the actual effective computing power of the server due to the single selected computing power component in the prior art.
[0190] At the same time, the technical solution disclosed in the present invention can solve the problem of inaccurate determination of the actual effective computing power of the server due to the difference between the test set and the actual tasks performed by the server by collecting the actual operating parameters generated during the actual operation of the server to determine the actual effective computing power of the server. At the same time, it can also avoid the occupation of server resources during the test process.
[0191] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.
[0192] Corresponds to Figure 1 The method for determining the effective computing power of the server in the present disclosure also provides an electronic device 400, such as Figure 4 FIG. 4 is a schematic diagram of the structure of an electronic device 400 provided in an embodiment of the present disclosure, including:
[0193] Processor 41, memory 42, and bus 43; memory 42 is used to store execution instructions, including internal memory 421 and external memory 422; the internal memory 421 is also called internal memory, which is used to temporarily store the operation data in the processor 41 and the data exchanged with the external memory 422 such as a hard disk. The processor 41 exchanges data with the external memory 422 through the internal memory 421. When the electronic device 400 is running, the processor 41 communicates with the memory 42 through the bus 43, so that the processor 41 executes the following instructions:
[0194] Acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters;
[0195] Determine the effective computing power of each of the core computing power components based on the actual operating parameters;
[0196] Based on the effective computing power of each of the core computing power components, the normalized effective computing power of the core computing power component is determined, and based on the normalized effective computing power of all the core computing power components, the actual effective computing power of the server is determined.
[0197] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.
[0198] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including," "comprising," "having," and the like are open words, referring to "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.
[0199] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.
[0200] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.
[0201] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.
[0202] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.
[0203] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.
Claims
1. A method for determining the effective computing power of a server, characterized in that: include: Acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters; Determine the effective computing power of each of the core computing power components based on the actual operating parameters; Based on the effective computing power of each of the core computing power components, the normalized effective computing power of the core computing power component is determined, and based on the normalized effective computing power of all the core computing power components, the actual effective computing power of the server is determined.
2. The method according to claim 1, characterized in that The determining the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components includes: Determining operating parameters of each benchmark computing component of the same type as the core computing component in the benchmark server; Determine the nominal computing power of each of the benchmark computing power components based on the operating parameters; The normalized effective computing power of the core computing power component is determined according to the nominal computing power and the effective computing power of the core computing power component.
3. The method according to claim 1, characterized in that The determining the actual effective computing power of the server based on the normalized effective computing power of all the core computing power components includes: Based on the normalized effective computing power of each of the core computing power components, determine the normalized effective computing power of the server; wherein the normalized effective computing power of the server is used to indicate the multiple relationship between the effective computing power of the server and the nominal computing power of the benchmark server; Based on the normalized effective computing power of the server, the actual effective computing power of the server is determined.
4. The method according to claim 3, characterized in that The core computing power components include at least one of the following: a central processing unit, a graphics processing unit, a memory and a hard disk.
5. The method according to claim 4, characterized in that The determining the normalized effective computing power of the server based on the normalized effective computing power of each of the core computing power components includes: Performing weighted summation on the normalized effective computing power of the central processing unit and the normalized effective computing power of the graphics processing unit according to the multiplication factor to obtain a weighted summation result; Determine a weight coefficient of the weighted sum result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk; wherein the weight coefficient is used to indicate the importance of the corresponding core computing power component to the server; The weighted summation result, the normalized effective computing power of the memory, and the normalized effective computing power of the hard disk are subjected to exponential multiplication according to the weight coefficient to obtain a calculation result, and the normalized effective computing power of the server is determined according to the calculation result.
6. The method according to claim 1, characterized in that The core computing power component is a central processing unit; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes: Acquire a first dynamic operating parameter and a first static operating parameter of the central processor; wherein the first static operating parameter includes the total number of cores of the central processor of the server and the single-cycle floating-point computing capability of each core of the central processor of the server, and the first dynamic operating parameter includes the single-core main frequency of each core of the central processor of the server and the utilization rate of each core of the central processor of the server; The effective computing power of the central processing unit is calculated according to the first dynamic operating parameter and the first static operating parameter.
7. The method according to claim 1, characterized in that The core computing power component is a graphics processor; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes: Acquire a second dynamic operating parameter and a second static operating parameter of the graphics processor; wherein the second static operating parameter includes the total number of cores of the graphics processor of the server and the single-cycle floating-point computing capability of each core of the graphics processor of the server, and the second dynamic operating parameter includes the single-core main frequency of each core of the graphics processor of the server and the utilization rate of each core of the graphics processor of the server; The effective computing power of the graphics processor is calculated according to the second dynamic operating parameter and the second static operating parameter.
8. The method according to claim 1, characterized in that The core computing power component is a memory; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes: Obtaining a third dynamic operating parameter and a third static operating parameter of the memory; wherein the third static operating parameter includes the number of memory bars of the server, the memory capacity of each memory bar of the server, the memory operating frequency of the server, the number of memory channels of the server, the bit width of the memory channel of the server, and the conversion coefficient of the memory configuration of each memory channel of the server to the memory computing power; and the third dynamic operating parameter includes the memory usage rate of each memory bar of the server; The effective computing power of the memory is calculated according to the third dynamic operating parameter and the third static operating parameter.
9. The method according to claim 1, characterized in that The core computing power component is a hard disk; and determining the effective computing power of each of the core computing power components based on the actual operating parameters includes: Acquire a fourth dynamic operating parameter and a fourth static operating parameter of the hard disk; wherein the fourth static operating parameter includes the number of hard disks of the server and the hard disk capacity of each hard disk of the server, and the fourth dynamic operating parameter includes the read and write speed of each hard disk of the server and the utilization rate of each hard disk of the server; The effective computing power of the hard disk is calculated according to the fourth dynamic operating parameter and the fourth static operating parameter.
10. A device for determining effective computing power of a server, characterized in that: include: An acquisition module, used to acquire actual operating parameters of each core computing component of the server during actual operation; wherein the actual operating parameters include dynamic operating parameters and static operating parameters; A first determination module, configured to determine the effective computing power of each of the core computing power components based on the actual operating parameters; The second determination module is used to determine the normalized effective computing power of the core computing power component based on the effective computing power of each of the core computing power components, and to determine the actual effective computing power of the server based on the normalized effective computing power of all the core computing power components.
11. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.
12. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method as claimed in any one of claims 1 to 9 are implemented.
13. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method as claimed in any one of claims 1 to 9 are implemented.
